Drivers´ en-route divergence behavior modeling using Extended Belief-Desire-Intention (E-BDI) framework

The goal of this paper is to analyze drivers´ en-route divergence behaviors when a road way is blocked by a car incident. The Extended Belief-Desire-Intention (E-BDI) framework is adopted in this work to mimic real drivers´ uncertain en-route planning behaviors based on the drivers´ perceptions and experiences. The proposed approach is implemented in Java-based E-BDI modules and DynusT® traffic simulation software, where a traffic data of Phoenix in the U.S. is used to illustrate and demonstrate the proposed approach. For validation of the proposed approach, we compare the drivers´ en-route divergence patterns obtained by E-BDI en-route planning with the divergence patterns provided by Time Dependent Shortest Path (TDSP) finding algorithm of DynusT®. The results have revealed that the proposed approach allows us to better understand various divergence patterns of drivers so that a reliable traffic system considering impacts of the sudden road way blocking events can be designed.